170 On definite Proportion3. 
equal quantities of oxygen. If now we compute the quantity of 
sulphuric acid necessary for their neutralisation, there will remain 
a quantity of water of crystallization containing seven times as 
much oxygen as.either base. Consequently in this double salt, 
each of the two simple salts retains its appropriate water of:cry- 
stallization, as appears from the respective analyses. The quan- 
tities of oxygen of the several component parts are here related, 
as in the former case, in the proportions of 1, 1, 6, and 7. 
a Alum. 
Alum is generally considered as a double ii but my 
analysis appears to show that it is in fact a neutral double salt. 
Twenty grammes of pure alum were heated in a platina eru- 
cible, over a spirit lamp, till they lost no more of their weiglit. 
The swollen mass was compressed into the crucible, and at last 
covered, in order that the heat might pervade every part of it 
eaually. It had now lost 9 gr. and [ could find no trace of the 
escape of an acid, which must have been discoverable by the 
smell. Consequently alum contains 45 per cent. of water of ery- 
stallization. The dry salt was again dissolved in water, the as- 
sistance of heat being required for the solution; muriate of 
baryta was added, and “the precipitate, when ignited, amounted 
to 19-973 gr. or very nearly to the original weight of the alum: 
so that, in a crystallized state, this substance contains 34°255 
per cent. of sulphuric acid [; or, according to the corrected 
analysis of the sulphate of baryta, which appears to contain from 
34-314 to 34:48 per cent. of sulphuric acid, at least 34:27 per 
cent. Gillert.]. 
Ten grammes of alum were dissolved in water, and digested 
with an excess of ammonia: they afforded alumina, which, when 
well washed and burnt, amounted to 10°67 per cent. As this did 
not agree with Mr. Thenard’s result, I repeated the experiment 
with 50 gr. of alum. After filtration, the fluid mixed with the 
excess of ammonia, and the water, with which the earth was 
washed, were evaporated to dryness, and redissolved in water, 
when they afforded a small quantity more of alumina: the whole, 
being collected and ignited, weighed 5:43 gr. and lost nothing 
“-more of its weight by repeated ignition. Consequently alum 
contains only from 1067 to 10:86 per cent. of alumina, 
Ten grammes of alum, dissolved in water, and digested in a 
small glass flask with carbonate of strontia, as long as any effer~ 
vescence existed, and then with new portions of the same car- 
bonate added in excess, were deprived in this manneryof the 
whole of the sulphate of alumina. The precipitate was not swol- 
len, as might have been expected from the alumina, but heavy 
and easily washed. ‘The fluid, when filtered, was not alkaline, 
mor 
